International Concrete Abstracts Portal

Force-deformation relationship of high-strength reinforced concrete beam members observed in the laboratory test was idealized by a trilinear relation for use in a nonlinear earthquake response analysis. Methods to evaluate the relationship were examined and the reliability of the methods were discussed with respect to the observed relations. Calculated initial stiffness is shown to significantly underestimate the observed value; a large coefficient of variation was attributed to accidental and shrinkage cracking in the specimen prior to the test. A similar large coefficient of variation was observed in the evaluation of cracking moment. Yield and ultimate moments could be favorably estimated by the theory. An empirical formula was proposed to evaluate yield deformation. An importance of controlling the elastic modulus of concrete in construction is emphasized if a structure is expected to behave as designed during an earthquake.

Objective evaluation of structural damage in buildings which have been subjected to strong ground motions is an undertaking in which expert knowledge and the ability to process correlated but fuzzy information in a consistent way must be blended. Often, in the immediate aftermath of earthquakes, field data is collected by survey teams whose expertise is variable. The use of knowledge-based systems capable of reaching an unequivocal decision on the damage state of a given building on the basis of queries arranged in a consistent hierarchical order would remove human subjectivity. This paper describes the internal design of an expert system called EPEDA, which is used as a tool for making a numerical ranking of damage in reinforced concrete buildings. Damage to individual elements is quantified on the basis of severity, relative member importance, and number of affected elements. Factors contributory in nature to the damage are summed with this score, as are scores expressing the overall system vulnerability. The final score is expressed as a number ranging from zero to 100. An example case is worked out to illustrate how the system works.

Following the strong earthquake in Chile on March 3, 1985, an intensive study was conducted to ascertain why the large inventory of moderate rise buildings in the coastal city of Vina del Mar performed so well during the earthquake. The major findings were that the vast majority of the buildings in this coastal city had a high wall area to total floor area ratio and that the reinforcement detailing in the boundaries of these walls were considerably less than required by U. S. codes. Analytical studies indicated that the high percentage of walls led to significantly lower drifts under severe seismic shaking, thus lowering the ductility demands on the walls. At lower levels of ductility demand, experimental results have demonstrated that wall boundaries did not need special detailing of transverse reinforcement. The findings from the series of research studies following the Chilean earthquake have led to modified U. S. design procedures that relate the need for special detailing in wall boundary elements to expected strain levels along the compression edge of the wall. The expected strain levels are determined based on the aspect ratio of the wall and the percentage of wall area to floor area used in the building.

An experimental program was carried out to investigate the behavior of concrete filled steel plate walls. Seven wall-panel specimens were tested under repetitive in-plane pure shear loading. Each specimen was made by connecting a pair of surface steel plates with partitioning webs and tie bars, and filling the boxes so-formed with concrete. The parameters investigated were the thickness of the surface steel plate, the number of partitioning webs and the presence or absence of headed stud bolts. Results describing a restoration force characteristic of a large loop area are presented. Rigidity after the onset of cracking approximates the cumulative value of truss rigidity (rigidity of resistance mechanism consisting of longitudinal and transverse tension chord members of steel plates and compression diagonals of concrete) and in-plane shearing rigidity of surface steel plates. The skeleton curve for the shear stress vs. shear strain relationship could be theoretically idealized into a quadri-linear curve with three control points.

Describes a research program to investigate the behavior of ductile connections between precast beam-column elements. Eight beam-column connections were tested to characterize the overall behavior of the connection details. Each connection specimen was designed to incorporate one of three behavioral concepts in the connection elements: tension/compression yielding, substantial energy dissipation, or nonlinear-elastic response. Based on the behavioral information collected during connection tests, analytical models were developed to investigate the behavior of complete precast frame systems. Results of the experimental study and preliminary results of the analytical work are presented. The objective of the program is to provide rational design recommendations for engineers to detail precast frame connections for use in regions of seismic risk.